Skip to content

P
Postgres FD Implementation
Commit 9b22846e authored by Bruce Momjian's avatar Bruce Momjian
Browse files
Options

Update docs for release.

parent 1561684a
Hide whitespace changes
Inline Side-by-side
Showing with 1030 additions and 919 deletions
doc/FAQ
View file @ 9b22846e
<HTML>
<HEAD>
<TITLE>PostgreSQL FAQ</title>
</HEAD>
<BODY BGCOLOR="#FFFFFF" TEXT="#000000" LINK="#FF0000" VLINK="#A00000" ALINK="#0000FF">
<H1>
Frequently Asked Questions (FAQ) for PostgreSQL
</H1>
<P>
Last updated: Wed Jun 11 10:44:40 EDT 1997
<BR>
Version: 6.1
<P>
Current maintainer: Bruce Momjian (<a
href="mailto:maillist@candle.pha.pa.us">maillist@candle.pha.pa.us</a>)<BR>
<P>
The most recent version of this document can be viewed at
the postgreSQL Web site, <a
href="http://postgreSQL.org">http://postgreSQL.org</a>.
<P>
Linux-specific questions are answered in
<a href="http://postgreSQL.org/docs/FAQ-Linux.phtml">http://postgreSQL.org/docs/FAQ-Linux.phtml</a>.
<P>
Irix-specific questions are answered in
<a href="http://postgreSQL.org/docs/FAQ-Irix.phtml">http://postgreSQL.org/docs/FAQ-Irix.phtml</a>.
<P>
Changes in this version (* = modified, + = new):
<UL>
<LI>3.42) What is Genetic Query Optimization?<BR>
<LI>3.43) I am running Solaris and my dates display wrong. Why?<BR>
</UL>
<HR>
<P>
<H2>Questions answered:</H2>
<H3> 1) General questions</H3>
<a href="#1.1">1.1</a>) What is PostgreSQL?<BR>
<a href="#1.2">1.2</a>) What does PostgreSQL run on?<BR>
<a href="#1.3">1.3</a>) Where can I get PostgreSQL?<BR>
<a href="#1.4">1.4</a>) What's the copyright on PostgreSQL?<BR>
<a href="#1.5">1.5</a>) Support for PostgreSQL<BR>
<a href="#1.6">1.6</a>) Latest release of PostgreSQL<BR>
<a href="#1.7">1.7</a>) Is there a commercial version of PostgreSQL?<BR>
<a href="#1.9">1.9</a>) What version of SQL does PostgreSQL use?<BR>
<a href="#1.10">1.10</a>) Does PostgreSQL work with databases from
earlier versions of postgres?<BR>
<a href="#1.11">1.11</a>) How many people use PostgreSQL?<BR>
<H3> 2) Installation questions</H3>
<a href="#2.1">2.1</a>) initdb doesn't run<BR>
<a href="#2.2">2.2</a>) when I start up the postmaster, I get
"FindBackend: could not find a backend to execute..."
"postmaster: could not find backend to execute..."<BR>
<a href="#2.3">2.3</a>) The system seems to be confused about commas,
decimal points, and date formats.<BR>
<a href="#2.4">2.4</a>) How do I install PostgreSQL somewhere other than
/usr/local/pgsql?<BR>
<a href="#2.5">2.5</a>) When I run postmaster, I get a Bad System Call
core dumped message.<BR>
<a href="#2.6">2.6</a>) When I try to start the postmaster, I get
IpcMemoryCreate errors.<BR>
<a href="#2.7">2.7</a>) I have changed a source file, but a
recompile does not see the change?<BR>
<H3> 3) Operational questions</H3>
<a href="#3.1">3.1</a>) How do I specify a KEY or other constraints on a
column?<BR>
<a href="#3.2">3.2</a>) Does PostgreSQL support nested subqueries?<BR>
<a href="#3.3">3.3</a>) How do I define a unique indices?<BR>
<a href="#3.4">3.4</a>) I've having a lot of problems using rules.<BR>
<a href="#3.5">3.5</a>) I can't seem to write into the middle of large
objects reliably.<BR>
<a href="#3.6">3.6</a>) Does PostgreSQL have a graphical user interface?
A report generator? A embedded query language interface?<BR>
<a href="#3.7">3.7</a>) How can I write client applications to
PostgreSQL?<BR>
<a href="#3.8">3.8</a>) How do I prevent other hosts from accessing my
PostgreSQL<BR>
<a href="#3.9">3.9</a>) How do I set up a pg_group?<BR>
<a href="#3.10">3.10</a>) What is the exact difference between
binary cursors and normal cursors?<BR>
<a href="#3.11">3.11</a>) Why doesn't the != operator work?<BR>
<a href="#3.12">3.12</a>) What is a R-tree index and what is it
used for?<BR>
<a href="#3.13">3.13</a>) What is the maximum size for a
tuple?<BR>
<a href="#3.14">3.14</a>) I defined indices but my queries don't
seem to make use of them. Why?<BR>
<a href="#3.15">3.15</a>) Are there ODBC drivers for
PostgreSQL?<BR>
<a href="#3.16">3.16</a>) How do I use postgres for
multi-dimensional indexing (&gt; 2 dimensions)?<BR>
<a href="#3.17">3.17</a>) How do I do regular expression searches?
case-insensitive regexp searching?<BR>
<a href="#3.18">3.18</a>) I can't access the database as the
'root' user.<BR>
<a href="#3.19">3.19</a>) I experienced a server crash during a
vacuum. How do I remove the lock file?<BR>
<a href="#3.20">3.20</a>) What is the difference between the
various character types?<BR>
<a href="#3.21">3.21</a>) In a query, how do I detect if a field
is NULL?<BR>
<a href="#3.22">3.22</a>) How do I see how the query optimizer is
evaluating my query?<BR>
<a href="#3.23">3.23</a>) How do I create a serial field?<BR>
<a href="#3.24">3.24</a>) How do I create a multi-column
index?<BR>
<a href="#3.25">3.25</a>) What are the temp_XXX files in my
database directory?<BR>
<a href="#3.26">3.26</a>) Why are my table files not getting any
smaller after a delete?<BR>
<a href="#3.27">3.27</a>) Why can't I connect to my database from
another machine?<BR>
<a href="#3.28">3.28</a>) I get the error 'default index class
unsupported' when creating an index. How do I do it?<BR>
<a href="#3.29">3.29</a>) Why does creating an index crash the
backend server?<BR>
<a href="#3.30">3.30</a>) How do I find out what indexes or
operations are defined in the database?<BR>
<a href="#3.31">3.31</a>) Why do statements require an extra character at
the end? Why does 'createuser' return 'unexpected last match in
input()'? Why does pg_dump fail?<BR>
<a href="#3.32">3.32</a>) All my servers crash under concurrent
table access. Why?<BR>
<a href="#3.33">3.33</a>) What tools are available for hooking
postgres to Web pages?<BR>
<a href="#3.34">3.34</a>) What is the time-warp feature and how
does it relate to vacuum?<BR>
<a href="#3.35">3.35</a>) How do I tune the database engine for
better performance?<BR>
<a href="#3.36">3.36</a>) What debugging features are available in
PostgreSQL?<BR>
<a href="#3.37">3.37</a>) What is an oid? What is a tid?<BR>
<a href="#3.38">3.38</a>) What is the meaning of some of the terms
used in Postgres?<BR>
<a href="#3.39">3.39</a>) What is Genetic Query Optimization?<BR>
<a href="#3.40">3.40</a>) I am running Solaris and my dates
display wrong. Why?
<a href="#3.41">3.41</a>) How do I enable more than 32 concurrent
backends?
<H3> 4) Questions about extending PostgreSQL</H3>
<a href="#4.1">4.1</a>) I wrote a user-defined function and when I run
it in psql, it dumps core.<BR>
<a href="#4.2">4.2</a>) I get messages of the type
NOTICE:PortalHeapMemoryFree: 0x402251d0<BR>
<a href="#4.3">4.3</a>) I've written some nifty new types and functions
for PostgreSQL.<BR>
<a href="#4.4">4.4</a>) How do I write a C function to return a
tuple?<BR>
<H3> 5) Bugs</H3>
<a href="#5.1">5.1</a>) How do I make a bug report?
Frequently Asked Questions (FAQ) for PostgreSQL <HR> <H2> Section 1: General Questions</H2> <H3><a
name="1.1">1.1</a>) What is PostgreSQL?</H3>
Last updated: Tue May 13 23:15:59 EDT 1997
Version: 6.1
Current maintainer: Bruce Momjian (maillist@candle.pha.pa.us)
The most recent version of this document can be viewed at the
postgreSQL Web site, http://postgreSQL.org.
Linux-specific questions are answered in
http://postgreSQL.org/docs/FAQ-Linux.phtml.
Irix-specific questions are answered in
http://postgreSQL.org/docs/FAQ-Irix.phtml.
Changes in this version (* = modified, + = new):
* 3.42) What is Genetic Query Optimization?
* 3.43) I am running Solaris and my dates display wrong. Why?
_________________________________________________________________
Questions answered:
1) General questions <P>
PostgreSQL is an enhancement of the POSTGRES database management system,
1.1) What is PostgreSQL? a next-generation DBMS research prototype. While PostgreSQL retains the
1.2) What does PostgreSQL run on? powerful data model and rich data types of POSTGRES, it replaces the
1.3) Where can I get PostgreSQL? PostQuel query language with an extended subset of SQL. PostgreSQL is
1.4) What's the copyright on PostgreSQL? free and the complete source is available.
1.5) Support for PostgreSQL <P>
1.6) Latest release of PostgreSQL PostgreSQL development is being performed by a team of Internet
1.7) Is there a commercial version of PostgreSQL? developers who all subscribe to the PostgreSQL development mailing list.
1.9) What version of SQL does PostgreSQL use? The current coordinator is Marc G. Fournier (<a
1.10) Does PostgreSQL work with databases from earlier versions of href="mailto:scrappy@postgreSQL.org">scrappy@postgreSQL.org</a>). (See
postgres? below on how to join). This team is now responsible for all current and
1.11) How many people use PostgreSQL? future development of PostgreSQL.
<P>
2) Installation questions The authors of PostgreSQL 1.01 were Andrew Yu and Jolly Chen. Many
others have contributed to the porting, testing, debugging and
2.1) initdb doesn't run enhancement of the code. The original Postgres code, from which
2.2) when I start up the postmaster, I get "FindBackend: could not PostgreSQL is derived, was the effort of many graduate students,
find a backend to execute..." "postmaster: could not find backend to undergraduate students, and staff programmers working under the
execute..." direction of Professor Michael Stonebraker at the University of
2.3) The system seems to be confused about commas, decimal points, and California, Berkeley.
date formats. <P>
2.4) How do I install PostgreSQL somewhere other than The original name of the software at Berkeley was Postgres. When SQL
/usr/local/pgsql? functionality was added in 1995, its name was changed to Postgres95. The
2.5) When I run postmaster, I get a Bad System Call core dumped name was changed at the end of 1996 to PostgreSQL.
message. <P>
2.6) I get the error message "obj/fmgr.h: No such file or directory" <H3><a name="1.2">1.2</a>) What does PostgreSQL run
2.7) When I try to start the postmaster, I get IpcMemoryCreate errors. on?</H3>
2.8) I have changed a source file, but a recompile does not see the <P>
change? The authors have compiled and tested PostgreSQL on the following
platforms(some of these compiles require gcc 2.7.0):
3) Operational questions <UL>
<LI> aix - IBM on AIX 3.2.5
3.1) How do I specify a KEY or other constraints on a column? <LI> alpha - DEC Alpha AXP on OSF/1 2.0
3.2) Does PostgreSQL support nested subqueries? <LI> BSD44_derived - OSs derived from 4.4-lite BSD (NetBSD, FreeBSD)
3.3) How do I define a unique indices? <LI> bsdi - BSD/OS 2.0, 2.01, 2.1, 3.0
3.4) I've having a lot of problems using rules. <LI> dgux - DG/UX 5.4R3.10
3.5) I can't seem to write into the middle of large objects reliably. <LI> hpux - HP PA-RISC on HP-UX 9.0
3.6) Does PostgreSQL have a graphical user interface? A report <LI> i386_solaris - i386 Solaris
generator? A embedded query language interface? <LI> irix5 - SGI MIPS on IRIX 5.3
3.7) How can I write client applications to PostgreSQL? <LI> linux - Intel x86 on Linux 1.2 and Linux ELF
3.8) How do I prevent other hosts from accessing my PostgreSQL (For non-ELF Linux, see LINUX_ELF below).
3.9) How do I set up a pg_group? <LI> sparc_solaris - SUN SPARC on Solaris 2.4
3.10) What is the exact difference between binary cursors and normal <LI> sunos4 - SUN SPARC on SunOS 4.1.3
cursors? <LI> svr4 - Intel x86 on Intel SVR4
3.11) Why doesn't the != operator work? <LI> ultrix4 - DEC MIPS on Ultrix 4.4
3.12) What is a R-tree index and what is it used for? </UL>
3.13) What is the maximum size for a tuple? The following platforms have known problems/bugs:
3.14) I defined indices but my queries don't seem to make use of them. <UL>
Why? <LI> nextstep - Motorola MC68K or Intel x86 on NeXTSTEP 3.2
3.15) Are there ODBC drivers for PostgreSQL? </UL>
3.16) How do I use postgres for multi-dimensional indexing (> 2 <P>
dimensions)? <H3><a name="1.3">1.3</a>) Where can I get PostgreSQL?</H3>
3.17) How do I do regular expression searches? case-insensitive regexp <P> The primary anonymous ftp site for PostgreSQL is:
searching? <UL>
3.18) I can't access the database as the 'root' user. <LI> <a
3.19) I experienced a server crash during a vacuum. How do I remove href="ftp://ftp.postgreSQL.org/pub">ftp://ftp.postgreSQL.org/pub</a>
the lock file? </UL>
3.20) What is the difference between the various character types? <P> A mirror site exists at:
3.21) In a query, how do I detect if a field is NULL? <UL>
3.22) How do I see how the query optimizer is evaluating my query? <LI> <a
3.23) How do I create a serial field? href="ftp://postgres95.vnet.net/pub/postgres95">ftp://postgres95.vnet.net/pub/postgres95</a>
3.24) How do I create a multi-column index? <LI> <a
3.25) What are the temp_XXX files in my database directory? href="ftp://ftp.luga.or.at/pub/postgres95">ftp://ftp.luga.or.at/pub/postgres95</a>
3.26) Why are my table files not getting any smaller after a delete? <LI> <a
3.27) Why can't I connect to my database from another machine? href="ftp://cal011111.student.utwente.nl/pub/postgres95">ftp://cal011111.student.utwente.nl/pub/postgres95</a>
3.28) I get the error 'default index class unsupported' when creating <LI> <a
an index. How do I do it? href="ftp://ftp.uni-trier.de/pub/database/rdbms/postgres/postgres95">ftp://ftp.uni-trier.de/pub/database/rdbms/postgres/postgres95</a>
3.29) Why does creating an index crash the backend server? <LI> <a
3.30) How do I specify a decimal constant as a float8, or a string as href="ftp://rocker.sch.bme.hu">ftp://rocker.sch.bme.hu</a>
a text? Why am I getting poor precision? </UL>
3.31) How do I find out what indexes or operations are defined in the <H3><a name="1.4">1.4</a>) What's the copyright on
database? PostgreSQL?</H3>
3.32) My database is corrupt. I can't do anything. What should I do? <P>
3.33) Createdb, destroydb, createuser, destroyuser don't run. Why? PostgreSQL is subject to the following COPYRIGHT.
3.34) Why do statements require an extra character at the end? Why <P>
does 'createuser' return 'unexpected last match in input()'? Why does PostgreSQL Data Base Management System
pg_dump fail? <P>
3.35) All my servers crash under concurrent table access. Why? Copyright (c) 1994-6 Regents of the University of California
3.36) What tools are available for hooking postgres to Web pages? <P>
3.37) What is the time-warp feature and how does it relate to vacuum? Permission to use, copy, modify, and distribute this software and its
3.38) How do I tune the database engine for better performance? documentation for any purpose, without fee, and without a written
3.39) What debugging features are available in PostgreSQL? agreement is hereby granted, provided that the above copyright notice
3.40) What is an oid? What is a tid? and this paragraph and the following two paragraphs appear in all
3.41) What is the meaning of some of the terms used in Postgres? copies.
3.42) What is Genetic Query Optimization? <P>
3.42) I am running Solaris and my dates display wrong. Why? IN NO EVENT SHALL THE UNIVERSITY OF CALIFORNIA BE LIABLE TO ANY PARTY
FOR DIRECT, INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES,
4) Questions about extending PostgreSQL INCLUDING LOST PROFITS, ARISING OUT OF THE USE OF THIS SOFTWARE AND ITS
DOCUMENTATION, EVEN IF THE UNIVERSITY OF CALIFORNIA HAS BEEN ADVISED OF
4.1) I wrote a user-defined function and when I run it in psql, it THE POSSIBILITY OF SUCH DAMAGE.
dumps core. <P>
4.2) I get messages of the type NOTICE:PortalHeapMemoryFree: THE UNIVERSITY OF CALIFORNIA SPECIFICALLY DISCLAIMS ANY WARRANTIES,
0x402251d0 INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY
4.3) I've written some nifty new types and functions for PostgreSQL. AND FITNESS FOR A PARTICULAR PURPOSE. THE SOFTWARE PROVIDED HEREUNDER
4.4) How do I write a C function to return a tuple? IS ON AN "AS IS" BASIS, AND THE UNIVERSITY OF CALIFORNIA HAS NO
OBLIGATIONS TO PROVIDE MAINTENANCE, SUPPORT, UPDATES, ENHANCEMENTS, OR
5) Bugs MODIFICATIONS.
<P>
5.1) How do I make a bug report? <H3><a name="1.5">1.5</a>) Support for PostgreSQL </H3>
_________________________________________________________________ <P>
There is no official support for PostgreSQL from the original
Section 1: General Questions maintainers or from University of California, Berkeley. It is
maintained through volunteer effort only.
<P>
The main mailing list is: <a
href="mailto:questions@postgreSQL.org">questions@postgreSQL.org</a>. It
is available for discussion o f matters pertaining to PostgreSQL,
including but not limited to bug reports and fixes. For info on how to
subscribe, send a mail with the lines in the body (not the subject line)
<PRE>
<CODE>
subscribe
end
</CODE>
</PRE>
<P>
to <a
href="mailto:questions-request@postgreSQL.org">questions-request@postgreSQL.org</a>.
<P>
There is also a digest list available. To subscribe to this list, send
email to:
<a
href="mailto:questions-digest-request@postgreSQL.org">
questions-digest-request@postgreSQL.org</a> with a BODY of:
<PRE>
<KBD>
subscribe
end
</KBD>
</PRE>
Digests are sent out to members of this list whenever the main list has
received around 30k of messages.
<P>
There is a bugs mailing list available. To subscribe to this
list, send email to <a
href="mailto:bugs-request@postgreSQL.org">bugs-request@postgreSQL.org</a>
with a BODY of:
<P>
There is also a developers discussion mailing list available. To subscribe to this
list, send email to <a
href="mailto:hackers-request@postgreSQL.org">hackers-request@postgreSQL.org</a>
with a BODY of:
<P>
<PRE>
<KBD>
subscribe
end
</KBD>
</PRE>
<P>
Additional information about PostgreSQL can be found via the PostgreSQL
WWW home page at:
<BLOCKQUOTE>
<a
href="http://postgreSQL.org">http://postgreSQL.org</a>
</BLOCKQUOTE>
<P>
<H3><a name="1.6">1.6</a>) Latest release of PostgreSQL</H3>
<P>
The latest release of PostgreSQL is version 6.0, which was released on
January 31, 1997. 6.1 is scheduled for release soon. For information
about what is new in 6.1, see our TODO list on our WWW page.
<P>
We expect a 7.0 release in several months that will remove time-travel
and reduce by 50% the size of on-disk system columns maintained for each
row in a table. This release will also require a dump and restore.
<P>
<H3><a name="1.7">1.7</a>) Is there a commercial version of PostgreSQL?</H3>
<P>
Illustra Information Technology (a wholly owned subsidiary of Informix
Software, Inc.) sells an object-relational DBMS called Illustra that was
originally based on postgres. Illustra has cosmetic similarities to
PostgreSQL but has more features, is more robust, performs better, and
offers real documentation and support. On the flip side, it costs
money. For more information, contact <a
href="mailto:sales@illustra.com">sales@illustra.com</a>
<P>
<H3><a name="1.8">1.8</a>) What documentation is available for PostgreSQL?</H3>
<P>
A user manual, manual pages, and some small test examples are included
in the distribution. The sql and built-in manual pages are particularly
important.
<P>
The www page contains pointers to an implementation guide and five
papers written about postgres design concepts and features.
<P>
<H3><a name="1.9">1.9</a>) What version of SQL does PostgreSQL use?</H3>
<P>
PostgreSQL supports a subset of SQL-92. It has most of the important
constructs but lacks some of the functionality. The most visible
differences are:
<UL>
<LI> no support for nested subqueries
<LI> no HAVING clause under a GROUP BY
</UL>
<P>
On the other hand, you get to create user-defined types, functions,
inheritance etc. If you're willing to help with PostgreSQL coding,
eventually we can also add the missing features listed above.
<P>
<H3><a name="1.10">1.10</a>) Does PostgreSQL work with databases from
earlier versions of postgres?</H3>
<P>
PostgreSQL v1.09 is compatible with databases created with v1.01. Those
upgrading from 1.0 should read the directions in the
MIGRATION_1.0_TO_1.02 directory.
<P>
Upgrading to 6.0 requires a dump and restore from previous releases.
<P>
Upgrading to 6.1 requires a dump and restore from previous releases.
<P>
Those ugrading from versions earlier than 1.09 must upgrade to 1.09
first without a dump/reload, then dump the data from 1.09, and then load
it into 6.0 or 6.1.
<P>
<H3><a name="1.11">1.11</a>) How many people use PostgreSQL?</H3>
<P>
Since we don't have any licensing or registration scheme, it's
impossible to tell. We do know hundreds copies of PostgreSQL v1.* have
been downloaded, and that there many hundreds of subscribers to the
mailing lists.
<P>
<HR>
<H2> Section 2: Installation Questions
</H2>
<P>
<H3><a name="2.1">2.1</a>) initdb doesn't run</H3>
<P>
<UL>
<LI> check to see that you have the proper paths set
<LI> check that the 'postgres' user owns all the right files
<LI> ensure that there are files in $PGDATA/files, and that they
are non-empty. If they aren't, then "gmake install" failed for
some reason
</UL>
<P>
<H3><a name="2.2">2.2</a>) when I start up the postmaster, I get
"FindBackend: could not find a backend to execute..."
"postmaster: could not find backend to execute..."</H3>
<P>
You probably do not have the right path set up. The 'postgres'
executable needs to be in your path.
<P>
<H3><a name="2.3">2.3</a>) The system seems to be confused about commas,
decimal points, and date formats.</H3>
<P>
Check your locale configuration. PostgreSQL uses the locale settings of
the user that ran the postmaster process. Set those accordingly for
your operating environment.
<P>
<H3><a name="2.4">2.4</a>) How do I install PostgreSQL somewhere other than
/usr/local/pgsql?</H3>
<P>
You need to edit Makefile.global and change POSTGRESDIR accordingly, or
create a Makefile.custom and define POSTGRESDIR there.
<P>
<H3><a name="2.5">2.5</a>) When I run postmaster, I get a Bad System
Call core dumped message.</H3>
<P>
It could be a variety of problems, but first check to see that you have
system V extensions installed on your kernel. PostgreSQL requires kernel
support for shared memory.
<P>
<H3><a name="2.6">2.6</a>) When I try to start the postmaster, I get
IpcMemoryCreate errors.</H3>
<P>
You either do not have shared memory configured properly in kernel or
you need to enlarge the shared memory available in the kernel. The
exact amount you need depends on your architecture and how many buffers
you configure postmaster to run with. For most systems, with default
buffer sizes, you need a minimum of ~760K.
<P>
<H3><a name="2.7">2.7</a>) I have changed a source file, but a
recompile does not see the change?</H3>
<P>
The Makefiles do not have the proper dependencies for include files. You
have to do a 'make clean' and then another 'make'.
<P>
<HR>
<H2> Section 3: PostgreSQL Features
</H2>
1.1) What is PostgreSQL? <P>
<H3><a name="3.1">3.1</a>) How do I specify a KEY or other constraints on a
PostgreSQL is an enhancement of the POSTGRES database management column?</H3>
system, a next-generation DBMS research prototype. While PostgreSQL <P>
retains the powerful data model and rich data types of POSTGRES, it Column constraints are not supported in PostgreSQL. As a consequence,
replaces the PostQuel query language with an extended subset of SQL. the system does not check for duplicates.
PostgreSQL is free and the complete source is available. <P>
Under 6.0, create a unique index on the column. Attempts to create
PostgreSQL development is being performed by a team of Internet duplicate of that column will report an error.
developers who all subscribe to the PostgreSQL development mailing <P>
list. The current coordinator is Marc G. Fournier <H3><a name="3.2">3.2</a>) Does PostgreSQL support nested subqueries?</H3>
(scrappy@postgreSQL.org). (See below on how to join). This team is now <P>
responsible for all current and future development of PostgreSQL. Subqueries are not implemented, but they can be simulated using sql
functions.
The authors of PostgreSQL 1.01 were Andrew Yu and Jolly Chen. Many <P>
others have contributed to the porting, testing, debugging and <H3><a name="3.3">3.3</a>) How do I define a unique indices?</H3>
enhancement of the code. The original Postgres code, from which <P>
PostgreSQL is derived, was the effort of many graduate students, PostgreSQL 6.0 supports unique indices.
undergraduate students, and staff programmers working under the <P>
direction of Professor Michael Stonebraker at the University of <H3><a name="3.4">3.4</a>) I've having a lot of problems using rules.</H3>
California, Berkeley. <P>
Currently, the rule system in PostgreSQL is mostly broken. It works
The original name of the software at Berkeley was Postgres. When SQL enough to support the view mechanism, but that's about it. Use
functionality was added in 1995, its name was changed to Postgres95. PostgreSQL rules at your own peril.
The name was changed at the end of 1996 to PostgreSQL. <P>
<H3><a name="3.5">3.5</a>) I can't seem to write into the middle of large
1.2) What does PostgreSQL run on? objects reliably.</H3>
<P>
The authors have compiled and tested PostgreSQL on the following The Inversion large object system in PostgreSQL is also mostly broken.
platforms(some of these compiles require gcc 2.7.0): It works well enough for storing large wads of data and reading them
* aix - IBM on AIX 3.2.5 back out, but the implementation has some underlying problems. Use
* alpha - DEC Alpha AXP on OSF/1 2.0 PostgreSQL large objects at your own peril.
* BSD44_derived - OSs derived from 4.4-lite BSD (NetBSD, FreeBSD) <P>
* bsdi - BSD/OS 2.0, 2.01, 2.1, 3.0 <H3><a name="3.6">3.6</a>) Does PostgreSQL have a graphical user interface?
* dgux - DG/UX 5.4R3.10 A report generator? A embedded query language interface?</H3>
* hpux - HP PA-RISC on HP-UX 9.0 <P>
* i386_solaris - i386 Solaris No. No. No. Not in the official distribution at least. Some users have
* irix5 - SGI MIPS on IRIX 5.3 reported some success at using 'pgbrowse' and 'onyx' as frontends to
* linux - Intel x86 on Linux 1.2 and Linux ELF (For non-ELF Linux, PostgreSQL. Several contributions are working on tk based frontend
see LINUX_ELF below). tools. Ask on the mailing list.
* sparc_solaris - SUN SPARC on Solaris 2.4 <P>
* sunos4 - SUN SPARC on SunOS 4.1.3 <H3><a name="3.7">3.7</a>) How can I write client applications to PostgreSQL?</H3>
* svr4 - Intel x86 on Intel SVR4 <P>
* ultrix4 - DEC MIPS on Ultrix 4.4 PostgreSQL supports a C-callable library interface called libpq as well
as a Tcl-based library interface called libtcl.
The following platforms have known problems/bugs: <P>
* nextstep - Motorola MC68K or Intel x86 on NeXTSTEP 3.2 Others have contributed a perl interface and a WWW gateway to
PostgreSQL. See the PostgreSQL home pages for more details.
1.3) Where can I get PostgreSQL? <P>
<H3><a name="3.8">3.8</a>) How do I prevent other hosts from accessing my
The primary anonymous ftp site for PostgreSQL is: PostgreSQL backend?</H3>
* ftp://ftp.postgreSQL.org/pub <P>
Use host-based authentication by modifying the file $PGDATA/pg_hba
A mirror site exists at: accordingly.
* ftp://postgres95.vnet.net/pub/postgres95 <P>
* ftp://ftp.luga.or.at/pub/postgres95 <H3><a name="3.9">3.9</a>) How do I set up a pg_group?</H3>
* ftp://cal011111.student.utwente.nl/pub/postgres95 <P>
* ftp://ftp.uni-trier.de/pub/database/rdbms/postgres/postgres95 Currently, there is no easy interface to set up user groups. You have to
* ftp://rocker.sch.bme.hu explicitly insert/update the pg_group table. For example:
<PRE>
1.4) What's the copyright on PostgreSQL? <CODE>
jolly=&gt; insert into pg_group (groname, grosysid, grolist)
PostgreSQL is subject to the following COPYRIGHT. jolly=&gt; values ('posthackers', '1234', '{5443, 8261}');
INSERT 548224
PostgreSQL Data Base Management System jolly=&gt; grant insert on foo to group posthackers;
CHANGE
Copyright (c) 1994-6 Regents of the University of California jolly=&gt;
</CODE>
Permission to use, copy, modify, and distribute this software and its </PRE>
documentation for any purpose, without fee, and without a written <P>
agreement is hereby granted, provided that the above copyright notice The fields in pg_group are:
and this paragraph and the following two paragraphs appear in all <UL>
copies. <LI> groname: the group name. This a char16 and should
be purely alphanumeric. Do not include underscores
IN NO EVENT SHALL THE UNIVERSITY OF CALIFORNIA BE LIABLE TO ANY PARTY or other punctuation.
FOR DIRECT, INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES, <LI> grosysid: the group id. This is an int4.
INCLUDING LOST PROFITS, ARISING OUT OF THE USE OF THIS SOFTWARE AND This should be unique for each group.
ITS DOCUMENTATION, EVEN IF THE UNIVERSITY OF CALIFORNIA HAS BEEN <LI> grolist: the list of pg_user id's that belong in the group.
ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. This is an int4[].
</UL>
THE UNIVERSITY OF CALIFORNIA SPECIFICALLY DISCLAIMS ANY WARRANTIES, <P>
INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF <H3><a name="3.10">3.10</a>) What is the exact difference between binary
MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE SOFTWARE cursors and normal cursors?</H3>
PROVIDED HEREUNDER IS ON AN "AS IS" BASIS, AND THE UNIVERSITY OF <P>
CALIFORNIA HAS NO OBLIGATIONS TO PROVIDE MAINTENANCE, SUPPORT, Normal cursors return data back in ASCII format. Since data is stored
UPDATES, ENHANCEMENTS, OR MODIFICATIONS. natively in binary format, the system must do a conversion to produce
the ASCII format. In addition, ASCII formats are often large in size
1.5) Support for PostgreSQL than binary format. Once the attributes come back in ASCII, often the
client application then has to convert it to a binary format to
There is no official support for PostgreSQL from the original manipulate it anyway.
maintainers or from University of California, Berkeley. It is <P>
maintained through volunteer effort only. Binary cursors give you back the data in the native binary
representation. Thus, binary cursors will tend to be a little faster
The main mailing list is: questions@postgreSQL.org. It is available since there's less overhead of conversion.
for discussion o f matters pertaining to PostgreSQL, including but not <P>
limited to bug reports and fixes. For info on how to subscribe, send a However, ASCII is architectural neutral whereas binary representation
mail with the lines in the body (not the subject line) can differ between different machine architecture. Thus, if your client
machine uses a different representation than you server machine, getting
back attributes in binary format is probably not what you want. Also, if
subscribe your main purpose is displaying the data in ASCII, then getting it back
end in ASCII will save you some effort on the client side.
<P>
to questions-request@postgreSQL.org. <H3><a name="3.11">3.11</a>) Why doesn't the != operator work?</H3>
<P>
There is also a digest list available. To subscribe to this list, send SQL specifies &lt;&gt; as the inequality operator, and that is what we
email to: questions-digest-request@postgreSQL.org with a BODY of: have defined for the built-in types.
<P>
In 6.0, != is equivalent to &lt;&gt;.
subscribe <P>
end <H3><a name="3.12">3.12</a>) What is a R-tree index and what is it used for?</H3>
<P>
Digests are sent out to members of this list whenever the main list An r-tree index is used for indexing spatial data. A hash index can't
has received around 30k of messages. handle range searches. A B-tree index only handles range searches in a
single dimension. R-tree's can handle multi-dimensional data. For
There is a bugs mailing list available. To subscribe to this list, example, if a R-tree index can be built on an attribute of type 'point',
send email to bugs-request@postgreSQL.org with a BODY of: the system can more efficient answer queries like select all points
within a bounding rectangle.
There is also a developers discussion mailing list available. To <P>
subscribe to this list, send email to hackers-request@postgreSQL.org The canonical paper that describes the original R-Tree design is:
with a BODY of: <P>
Guttman, A. "R-Trees: A Dynamic Index Structure for Spatial Searching."
Proc of the 1984 ACM SIGMOD Int'l Conf on Mgmt of Data, 45-57.
subscribe <P>
end You can also find this paper in Stonebraker's "Readings in Database
Systems"
Additional information about PostgreSQL can be found via the <P>
PostgreSQL WWW home page at: <H3><a name="3.13">3.13</a>) What is the maximum size for a tuple?</H3>
<P>
http://postgreSQL.org Tuples are limited to 8K bytes. Taking into account system attributes
and other overhead, one should stay well shy of 8,000 bytes to be on the
1.6) Latest release of PostgreSQL safe side. To use attributes larger than 8K, try using the large
objects interface.
The latest release of PostgreSQL is version 6.0, which was released on <P>
January 31, 1997. 6.1 is scheduled for release soon. For information Tuples do not cross 8k boundaries so a 5k tuple will require 8k of
about what is new in 6.1, see our TODO list on our WWW page. storage.
<P>
We expect a 7.0 release in several months that will remove time-travel <H3><a name="3.14">3.14</a>) I defined indices but my queries don't seem
and reduce by 50% the size of on-disk system columns maintained for to make use of them. Why?</H3>
each row in a table. This release will also require a dump and <P>
restore. PostgreSQL does not automatically maintain statistics. One has to make
an explicit 'vacuum' call to update the statistics. After statistics
1.7) Is there a commercial version of PostgreSQL? are updated, the optimizer has a better shot at using indices. Note
that the optimizer is limited and does not use indices in some
Illustra Information Technology (a wholly owned subsidiary of Informix circumstances (such as OR clauses).
Software, Inc.) sells an object-relational DBMS called Illustra that <P>
was originally based on postgres. Illustra has cosmetic similarities If the system still does not see the index, it is probably because you
to PostgreSQL but has more features, is more robust, performs better, have created an index on a field with the improper *_ops type. For
and offers real documentation and support. On the flip side, it costs example, you have created a CHAR(4) field, but have specified a char_ops
money. For more information, contact sales@illustra.com index type_class.
<P>
1.8) What documentation is available for PostgreSQL? See the create_index manual page for information on what type classes
are available. It must match the field type.
A user manual, manual pages, and some small test examples are included <P>
in the distribution. The sql and built-in manual pages are Postgres does not warn the user when the improper index is created.
particularly important. <P>
Indexes not used for ORDER BY operations.
The www page contains pointers to an implementation guide and five <P>
papers written about postgres design concepts and features. <H3><a name="3.15">3.15</a>) Are there ODBC drivers for PostgreSQL?</H3>
<P>
1.9) What version of SQL does PostgreSQL use? There are two ODBC drivers available, PostODBC and OpenLink ODBC.
<P>
PostgreSQL supports a subset of SQL-92. It has most of the important For all people being interested in PostODBC, there are now two mailing
constructs but lacks some of the functionality. The most visible lists devoted to the discussion of PostODBC. The mailing lists are:
differences are: <UL>
* no support for nested subqueries <LI> <a
* no HAVING clause under a GROUP BY href="mailto:postodbc-users@listserv.direct.net">postodbc-users@listserv.direct.
net</a>
On the other hand, you get to create user-defined types, functions, <LI> <a
inheritance etc. If you're willing to help with PostgreSQL coding, href="mailto:postodbc-developers@listserv.direct.net">postodbc-developers@listse
eventually we can also add the missing features listed above. rv.direct.net</a>
</UL>
1.10) Does PostgreSQL work with databases from earlier versions of postgres? <P>
these lists are ordinary majordomo mailing lists. You can subscribe by
PostgreSQL v1.09 is compatible with databases created with v1.01. sending a mail to:
Those upgrading from 1.0 should read the directions in the <UL>
MIGRATION_1.0_TO_1.02 directory. <LI> <a
href="mailto:majordomo@listserv.direct.net">majordomo@listserv.direct.net</a>
Upgrading to 6.0 requires a dump and restore from previous releases. </UL>
<P>
1.11) How many people use PostgreSQL? OpenLink ODBC is currently in beta under Linux. You can get it from <a
href="http://www.openlinksw.com/postgres.html">
Since we don't have any licensing or registration scheme, it's http://www.openlinksw.com/postgres.html</a>. It works with our standard
impossible to tell. We do know hundreds copies of PostgreSQL v1.* have ODBC client software so you'll have Postgres ODBC available on every
been downloaded, and that there many hundreds of subscribers to the client platform we support (Win, Mac, Unix, VMS).
mailing lists. <P>
We will probably be selling this product to people who need
_________________________________________________________________ commercial-quality support, but a freeware version will always be
available. Questions to <a
Section 2: Installation Questions href="mailto:postgres95@openlink.co.uk">postgres95@openlink.co.uk</a>.
<P>
2.1) initdb doesn't run <H3><a name="3.16">3.16</a>) How do I use postgres for multi-dimensional
indexing (&gt; 2 dimensions)?</H3>
* check to see that you have the proper paths set <P>
* check that the 'postgres' user owns all the right files Builtin R-Trees can handle polygons and boxes. In theory, R-trees can
* ensure that there are files in $PGDATA/files, and that they are be extended to handle higher number of dimensions. In practice,
non-empty. If they aren't, then "gmake install" failed for some extending R-trees require a bit of work and we don't currently have any
reason documentation on how to do it.
<P>
2.2) when I start up the postmaster, I get "FindBackend: could not find a <H3><a name="3.17">3.17</a>) How do I do regular expression searches?
backend to execute..." "postmaster: could not find backend to execute..." case-insensitive regexp searching?</H3>
<P>
You probably do not have the right path set up. The 'postgres' PostgreSQL supports the SQL LIKE syntax as well as more general regular
executable needs to be in your path. expression searching with the ~ operator. The !~ is the negated regexp
operator. ~* and !~* are the case-insensitive regular expression
2.3) The system seems to be confused about commas, decimal points, and date operators.
formats. <P>
<H3><a name="3.18">3.18</a>) I can't access the database as the 'root' user.</H3>
Check your locale configuration. PostgreSQL uses the locale settings <P>
of the user that ran the postmaster process. Set those accordingly for You should not create database users with user id 0(root). They will be
your operating environment. unable to access the database. This is a security precaution because
of the ability of any user to dynamically link object modules into the
2.4) How do I install PostgreSQL somewhere other than /usr/local/pgsql? database engine.
<P>
You need to edit Makefile.global and change POSTGRESDIR accordingly, <H3><a name="3.19">3.19</a>) I experienced a server crash during a
or create a Makefile.custom and define POSTGRESDIR there. vacuum. How do I remove the lock file?</H3>
<P>
2.5) When I run postmaster, I get a Bad System Call core dumped message. If the server crashes during a vacuum command, chances are it will leave
a lock file hanging around. Attempts to re-run the vacuum command
It could be a variety of problems, but first check to see that you result in
have system V extensions installed on your kernel. PostgreSQL requires <PRE>
kernel support for shared memory. <SAMP>
WARN:can't create lock file -- another vacuum cleaner running?
2.6) I get the error message "obj/fmgr.h: No such file or directory" </SAMP>
</PRE>
This indicates that you did not generate the file fmgr.h properly. <P>
Something failed in the running of the If you are sure that no vacuum is actually running, you can remove the
src/backend/utils/Gen_fmgrtab.sh script. Check to see the paths used file called "pg_vlock" in your database directory (which is
in that script is appropriate to your system. $PGDATA/base/&lt;dbName&gt;)
<P>
2.7) When I try to start the postmaster, I get IpcMemoryCreate errors. <H3><a name="3.20">3.20</a>) What is the difference between the various
character types?</H3>
You either do not have shared memory configured properly in kernel or <PRE>
you need to enlarge the shared memory available in the kernel. The
exact amount you need depends on your architecture and how many
buffers you configure postmaster to run with. For most systems, with
default buffer sizes, you need a minimum of ~760K.
2.8) I have changed a source file, but a recompile does not see the change?
The Makefiles do not have the proper dependencies for include files.
You have to do a 'make clean' and then another 'make'.
_________________________________________________________________
Section 3: PostgreSQL Features
3.1) How do I specify a KEY or other constraints on a column?
Column constraints are not supported in PostgreSQL. As a consequence,
the system does not check for duplicates.
Under 6.0, create a unique index on the column. Attempts to create
duplicate of that column will report an error.
3.2) Does PostgreSQL support nested subqueries?
Subqueries are not implemented, but they can be simulated using sql
functions.
3.3) How do I define a unique indices?
PostgreSQL 6.0 supports unique indices.
3.4) I've having a lot of problems using rules.
Currently, the rule system in PostgreSQL is mostly broken. It works
enough to support the view mechanism, but that's about it. Use
PostgreSQL rules at your own peril.
3.5) I can't seem to write into the middle of large objects reliably.
The Inversion large object system in PostgreSQL is also mostly broken.
It works well enough for storing large wads of data and reading them
back out, but the implementation has some underlying problems. Use
PostgreSQL large objects at your own peril.
3.6) Does PostgreSQL have a graphical user interface? A report generator? A
embedded query language interface?
No. No. No. Not in the official distribution at least. Some users have
reported some success at using 'pgbrowse' and 'onyx' as frontends to
PostgreSQL. Several contributions are working on tk based frontend
tools. Ask on the mailing list.
3.7) How can I write client applications to PostgreSQL?
PostgreSQL supports a C-callable library interface called libpq as
well as a Tcl-based library interface called libtcl.
Others have contributed a perl interface and a WWW gateway to
PostgreSQL. See the PostgreSQL home pages for more details.
3.8) How do I prevent other hosts from accessing my PostgreSQL backend?
Use host-based authentication by modifying the file $PGDATA/pg_hba
accordingly.
3.9) How do I set up a pg_group?
Currently, there is no easy interface to set up user groups. You have
to explicitly insert/update the pg_group table. For example:
jolly=> insert into pg_group (groname, grosysid, grolist)
jolly=> values ('posthackers', '1234', '{5443, 8261}');
INSERT 548224
jolly=> grant insert on foo to group posthackers;
CHANGE
jolly=>
The fields in pg_group are:
* groname: the group name. This a char16 and should be purely
alphanumeric. Do not include underscores or other punctuation.
* grosysid: the group id. This is an int4. This should be unique for
each group.
* grolist: the list of pg_user id's that belong in the group. This
is an int4[].
3.10) What is the exact difference between binary cursors and normal cursors?
Normal cursors return data back in ASCII format. Since data is stored
natively in binary format, the system must do a conversion to produce
the ASCII format. In addition, ASCII formats are often large in size
than binary format. Once the attributes come back in ASCII, often the
client application then has to convert it to a binary format to
manipulate it anyway.
Binary cursors give you back the data in the native binary
representation. Thus, binary cursors will tend to be a little faster
since there's less overhead of conversion.
However, ASCII is architectural neutral whereas binary representation
can differ between different machine architecture. Thus, if your
client machine uses a different representation than you server
machine, getting back attributes in binary format is probably not what
you want. Also, if your main purpose is displaying the data in ASCII,
then getting it back in ASCII will save you some effort on the client
side.
3.11) Why doesn't the != operator work?
SQL specifies <> as the inequality operator, and that is what we have
defined for the built-in types.
In 6.0, != is equivalent to <>.
3.12) What is a R-tree index and what is it used for?
An r-tree index is used for indexing spatial data. A hash index can't
handle range searches. A B-tree index only handles range searches in a
single dimension. R-tree's can handle multi-dimensional data. For
example, if a R-tree index can be built on an attribute of type
'point', the system can more efficient answer queries like select all
points within a bounding rectangle.
The canonical paper that describes the original R-Tree design is:
Guttman, A. "R-Trees: A Dynamic Index Structure for Spatial
Searching." Proc of the 1984 ACM SIGMOD Int'l Conf on Mgmt of Data,
45-57.
You can also find this paper in Stonebraker's "Readings in Database
Systems"
3.13) What is the maximum size for a tuple?
Tuples are limited to 8K bytes. Taking into account system attributes
and other overhead, one should stay well shy of 8,000 bytes to be on
the safe side. To use attributes larger than 8K, try using the large
objects interface.
Tuples do not cross 8k boundaries so a 5k tuple will require 8k of
storage.
3.14) I defined indices but my queries don't seem to make use of them. Why?
PostgreSQL does not automatically maintain statistics. One has to make
an explicit 'vacuum' call to update the statistics. After statistics
are updated, the optimizer has a better shot at using indices. Note
that the optimizer is limited and does not use indices in some
circumstances (such as OR clauses).
If the system still does not see the index, it is probably because you
have created an index on a field with the improper *_ops type. For
example, you have created a CHAR(4) field, but have specified a
char_ops index type_class.
See the create_index manual page for information on what type classes
are available. It must match the field type.
Postgres does not warn the user when the improper index is created.
Indexes not used for ORDER BY operations.
3.15) Are there ODBC drivers for PostgreSQL?
There are two ODBC drivers available, PostODBC and OpenLink ODBC.
For all people being interested in PostODBC, there are now two mailing
lists devoted to the discussion of PostODBC. The mailing lists are:
* postodbc-users@listserv.direct. net
* postodbc-developers@listse rv.direct.net
these lists are ordinary majordomo mailing lists. You can subscribe by
sending a mail to:
* majordomo@listserv.direct.net
OpenLink ODBC is currently in beta under Linux. You can get it from
http://www.openlinksw.com/postgres.html. It works with our standard
ODBC client software so you'll have Postgres ODBC available on every
client platform we support (Win, Mac, Unix, VMS).
We will probably be selling this product to people who need
commercial-quality support, but a freeware version will always be
available. Questions to postgres95@openlink.co.uk.
3.16) How do I use postgres for multi-dimensional indexing (> 2 dimensions)?
Builtin R-Trees can handle polygons and boxes. In theory, R-trees can
be extended to handle higher number of dimensions. In practice,
extending R-trees require a bit of work and we don't currently have
any documentation on how to do it.
3.17) How do I do regular expression searches? case-insensitive regexp
searching?
PostgreSQL supports the SQL LIKE syntax as well as more general
regular expression searching with the ~ operator. The !~ is the
negated regexp operator. ~* and !~* are the case-insensitive regular
expression operators.
3.18) I can't access the database as the 'root' user.
You should not create database users with user id 0(root). They will
be unable to access the database. This is a security precaution
because of the ability of any user to dynamically link object modules
into the database engine.
3.19) I experienced a server crash during a vacuum. How do I remove the lock
file?
If the server crashes during a vacuum command, chances are it will
leave a lock file hanging around. Attempts to re-run the vacuum
command result in
WARN:can't create lock file -- another vacuum cleaner running?
If you are sure that no vacuum is actually running, you can remove the
file called "pg_vlock" in your database directory (which is
$PGDATA/base/<dbName>)
3.20) What is the difference between the various character types?
Type Internal Name Notes Type Internal Name Notes
-------------------------------------------------- --------------------------------------------------
CHAR char 1 character } CHAR char 1 character }
...@@ -592,343 +676,368 @@ CHAR(#) bpchar blank padded to the specified fixed length ...@@ -592,343 +676,368 @@ CHAR(#) bpchar blank padded to the specified fixed length
VARCHAR(#) varchar size specifies maximum length, no padding VARCHAR(#) varchar size specifies maximum length, no padding
TEXT text length limited only by maximum tuple length TEXT text length limited only by maximum tuple length
BYTEA bytea variable-length array of bytes BYTEA bytea variable-length array of bytes
</PRE>
<P>
Remember, you need to use the internal name when creating indexes on
these fields or when doing other internal operations.
<P>
The last four types above are "varlena" types (i.e. the first four bytes
is the length, followed by the data). CHAR(#) and VARCHAR(#) allocate
the maximum number of bytes no matter how much data is stored in the
field. TEXT and BYTEA are the only character types that have variable
length on the disk.
<P>
<H3><a name="3.21">3.21</a>) In a query, how do I detect if a field is NULL?</H3>
<P>
PostgreSQL has two builtin keywords, "isnull" and "notnull" (note no
spaces). Version 1.05 and later and 6.* understand IS NULL and IS NOT
NULL.
<P>
<H3><a name="3.22">3.22</a>) How do I see how the query optimizer is
evaluating my query?</H3>
<P>
Place the word 'EXPLAIN' at the beginning of the query, for example:
<PRE>
<CODE>
EXPLAIN SELECT * FROM table1 WHERE age = 23;
</CODE>
</PRE>
<P>
<H3><a name="3.23">3.23</a>) How do I create a serial field?</H3>
<P>
Postgres does not allow the user to specifiy a user column as type
SERIAL. Instead, you can use each row's oid field as a unique value.
However, if you need to dump and reload the database, you need to be
using postgres version 1.07 or later or 6.* with pgdump's -o option or
COPY's WITH OIDS option to preserver the oids.
<P>
Another valid way of doing this is to create a function:
<PRE>
<CODE>
create table my_oids (f1 int4);
insert into my_oids values (1);
create function new_oid () returns int4 as
'update my_oids set f1 = f1 + 1; select f1 from my_oids; '
language 'sql';
</CODE>
</PRE>
<P>
then:
<PRE>
<CODE>
create table my_stuff (my_key int4, value text);
insert into my_stuff values (new_oid(), 'hello');
</CODE>
</PRE>
<P>
However, keep in mind there is a race condition here where one server
could do the update, then another one do an update, and they both could
select the same new id. This statement should be performed within a
transaction.
<P>
Sequences are implemented in 6.1
<P>
<H3><a name="3.24">3.24</a>) How do I create a multi-column index?</H3>
<P>
Remember, you need to use the internal name when creating indexes on In 6.0, you can not directly create a multi-column index using create
these fields or when doing other internal operations. index. You need to define a function which acts on the multiple columns,
then use create index with that function.
The last four types above are "varlena" types (i.e. the first four <P>
bytes is the length, followed by the data). CHAR(#) and VARCHAR(#) In 6.1, this feature is available.
allocate the maximum number of bytes no matter how much data is stored <P>
in the field. TEXT and BYTEA are the only character types that have <H3><a name="3.25">3.25</a>) What are the temp_XXX files in my database
variable length on the disk. directory?</H3>
<P>
3.21) In a query, how do I detect if a field is NULL? They are temp_ files generated by the query executor. For example, if a
sort needs to be done to satisfy an ORDER BY, some temp files are
PostgreSQL has two builtin keywords, "isnull" and "notnull" (note no generated as a result of the sort.
spaces). Version 1.05 and later and 6.* understand IS NULL and IS NOT <P>
NULL. If you have no transactions or sorts running at the time, it is safe to
delete the temp_ files.
3.22) How do I see how the query optimizer is evaluating my query? <P>
<H3><a name="3.26">3.26</a>) Why are my table files not getting any
Place the word 'EXPLAIN' at the beginning of the query, for example: smaller after a delete?</H3>
<P>
EXPLAIN SELECT * FROM table1 WHERE age = 23; If you run vacuum in pre-6.0, unused rows will be marked for reuse, but
the file blocks are not released.
3.23) How do I create a serial field? <P>
In 6.0, vacuum properly shrinks tables.
Postgres does not allow the user to specifiy a user column as type <P>
SERIAL. Instead, you can use each row's oid field as a unique value. <H3><a name="3.27">3.27</a>) Why can't I connect to my database from
However, if you need to dump and reload the database, you need to be another machine?</H3>
using postgres version 1.07 or later or 6.* with pgdump's -o option or <P>
COPY's WITH OIDS option to preserver the oids. The default configuration allows only connections from tcp/ip host
localhost. You need to add a host entry to the file pgsql/data/pg_hba.
Another valid way of doing this is to create a function: <P>
<H3><a name="3.28">3.28</a>) I get the error 'default index class
create table my_oids (f1 int4); unsupported' when creating an index. How do I do it?</H3>
insert into my_oids values (1); <P>
create function new_oid () returns int4 as You probably used:
'update my_oids set f1 = f1 + 1; select f1 from my_oids; ' <PRE>
language 'sql'; <CODE>
create index idx1 on person using btree (name);
then: </CODE>
</PRE>
create table my_stuff (my_key int4, value text); <P>
insert into my_stuff values (new_oid(), 'hello'); PostgreSQL indexes are extensible, and therefore in pre-6.0, you must
specify a class_type when creating an index. Read the manual page for
However, keep in mind there is a race condition here where one server create index (called create_index).
could do the update, then another one do an update, and they both <P>
could select the same new id. This statement should be performed Version 6.0, if you do not specify a class_type, it defaults to the
within a transaction. proper type for the column.
<P>
3.24) How do I create a multi-column index? <H3><a name="3.29">3.29</a>) Why does creating an index crash the
backend server?</H3>
In 6.0, you can not directly create a multi-column index using create <P>
index. You need to define a function which acts on the multiple You have probably defined an incorrect *_ops type class for the field
columns, then use create index with that function. you are indexing.
<P>
In 6.1, this feature is available. <H3><a name="3.30">3.30</a>) How do I find out what indexes or
operations are defined in the database?</H3>
3.25) What are the temp_XXX files in my database directory? <P>
Run the file pgsql/src/tutorial/syscat.source. It illustrates many of
They are temp_ files generated by the query executor. For example, if the 'select's needed to get information out of the database system
a sort needs to be done to satisfy an ORDER BY, some temp files are tables.
generated as a result of the sort. <P>
<H3><a name="3.31">3.31</a>) Why do statements require an extra character at
If you have no transactions or sorts running at the time, it is safe the end? Why does 'createuser' return 'unexpected last match in input()'?
to delete the temp_ files. Why does pg_dump fail?</H3>
<P>
3.26) Why are my table files not getting any smaller after a delete? You have compile postgres with flex version 2.5.3. There is bug in this
version of flex. Use flex version 2.5.2 or flex 2.5.4 instead. There
If you run vacuum in pre-6.0, unused rows will be marked for reuse, is a doc/README.flex file which will properly patch the flex 2.5.3
but the file blocks are not released. source code.
<P>
In 6.0, vacuum properly shrinks tables. <H3><a name="3.32">3.32</a>) All my servers crash under concurrent table
access. Why?</H3>
3.27) Why can't I connect to my database from another machine? <P>
This problem can be caused by a kernel that is not configured to support
The default configuration allows only connections from tcp/ip host semaphores.
localhost. You need to add a host entry to the file pgsql/data/pg_hba. <P>
<H3><a name="3.33">3.33</a>) What tools are available for hooking
3.28) I get the error 'default index class unsupported' when creating an postgres to Web pages?</H3>
index. How do I do it? <P>
For web integration, PHP/FI is an excellent interface. The URL for that
You probably used: is <a href="http://www.vex.net/php/">http://www.vex.net/php/</a>
<P>
create index idx1 on person using btree (name); PHP is great for simple stuff, but for more complex stuff, some still
use the perl interface and CGI.pm.
PostgreSQL indexes are extensible, and therefore in pre-6.0, you must <P>
specify a class_type when creating an index. Read the manual page for An example of using WWW with C to talk to Postgres is can be tried at:
create index (called create_index). <UL>
<LI>
Version 6.0, if you do not specify a class_type, it defaults to the <a href="http://postgreSQL.org/%7Emlc">http://postgreSQL.org/~mlc</a>
proper type for the column. </UL>
<P>
3.29) Why does creating an index crash the backend server? An WWW gatway based on WDB using perl can be downloaded from:
<UL>
You have probably defined an incorrect *_ops type class for the field <LI>
you are indexing. <a
href="http://www.eol.ists.ca/~dunlop/wdb-p95">http://www.eol.ists.ca/~dunlop/wdb
3.30) How do I specify a decimal constant as a float8, or a string as a text? -p95</a>
Why am I getting poor precision? </UL>
<H3><a name="3.34">3.34</a>) What is the time-warp feature and how does
Use the :: operator. It is needed only when the default promotion it relate to vacuum?</H3>
rules fail. i.e.: <P>
PostgreSQL handles data changes differently than most database systems.
When a row is changed in a table, the original row is marked with the
insert into tab1 values (4.23::float8, '2343'::text) time it was changed, and a new row is created with the current data. By
default, only current rows are used in a table. If you specify a
The default floating-point constant is a float4 in releases prior to date/time after the table name in a FROM clause, you can access the data
1.05. Later releases default to float8. that was current at that time, i.e.
<PRE>
3.31) How do I find out what indexes or operations are defined in the <CODE>
database? SELECT *
FROM employees ['July 24, 1996 09:00:00']
Run the file pgsql/src/tutorial/syscat.source. It illustrates many of </CODE>
the 'select's needed to get information out of the database system </PRE>
tables. <P>
displays employee rows in the table at the specified time. You can
3.32) My database is corrupt. I can't do anything. What should I do? specify intervals like [date,date], [date,], [,date], or [,]. This last
option accesses all rows that ever existed.
The 1.02 release has a README file and utility that describes a <P>
possible cause of the problem and a workaround. INSERTed rows get a timestamp too, so rows that were not in the table at
the desired time will not appear.
This bug is fixed in 1.02.1. <P>
Vacuum removes rows that are no longer current. This time-warp feature
3.33) Createdb, destroydb, createuser,destroyuser don't run. Why? is used by the engine for rollback and crash recovery. Expiration times
can be set with purge.
Release 1.02 does not have this problem. <P>
In 6.0, once a table is vacuumed, the creation time of a row may be
The 1.01 release of PostgreSQL uses a variable called PAGER to filter incorrect, causing time-traval to fail.
the output of SELECT statements. Unfortunately, this PAGER is used <P>
even when the standard output is not a terminal. The time-travel feature will be removed in 7.0.
<P>
3.34) Why do statements require an extra character at the end? Why does <H3><a name="3.35">3.35</a>) How do I tune the database engine for
'createuser' return 'unexpected last match in input()'? Why does pg_dump better performance?</H3>
fail? <P>
There are two things that can be done. You can use Openlink's option to
You have compile postgres with flex version 2.5.3. There is bug in disable fsync() by starting the postmaster with a '-o -F' option. This
this version of flex. Use flex version 2.5.2 or flex 2.5.4 instead. will prevent fsync()'s from flushing to disk after every transaction.
There is a doc/README.flex file which will properly patch the flex <P>
2.5.3 source code. You can also use the postmaster -B option to increase the number of
shared memory buffers shared among the backend processes. If you make
3.35) All my servers crash under concurrent table access. Why? this parameter too high, the process will not start or crash
unexpectedly. Each buffer is 8K and the defualt is 64 buffers.
This problem can be caused by a kernel that is not configured to <P>
support semaphores. <H3><a name="3.36">3.36</a>) What debugging features are available in
PostgreSQL?</H3>
3.36) What tools are available for hooking postgres to Web pages? <P>
PostgreSQL has several features that report status information that can
For web integration, PHP/FI is an excellent interface. The URL for be valuable for debugging purposes.
that is http://www.vex.net/php/ <P>
First, by compiling with DEBUG defined, many assert()'s monitor the
PHP is great for simple stuff, but for more complex stuff, some still progress of the backend and halt the program when something unexpected
use the perl interface and CGI.pm. occurs.
<P>
An example of using WWW with C to talk to Postgres is can be tried at: Both postmaster and postgres have several debug options available.
* http://postgreSQL.org/~mlc First, whenever you start the postmaster, make sure you send the
standard output and error to a log file, like:
An WWW gatway based on WDB using perl can be downloaded from: <PRE>
* http://www.eol.ists.ca/~dunlop/wdb -p95 <KBD>
cd /usr/local/pgsql
3.37) What is the time-warp feature and how does it relate to vacuum? ./bin/postmaster &gt;server.log 2&gt;&1 &
</KBD>
PostgreSQL handles data changes differently than most database </PRE>
systems. When a row is changed in a table, the original row is marked <P>
with the time it was changed, and a new row is created with the This will put a server.log file in the top-level PostgreSQL directory.
current data. By default, only current rows are used in a table. If This file can contain useful information about problems or errors
you specify a date/time after the table name in a FROM clause, you can encountered by the server. Postmaster has a -d option that allows even
access the data that was current at that time, i.e. more detailed information to be reported. The -d option takes a number
1-3 that specifies the debug level. The query plans in a verbose debug
file can be formatted using the 'indent' program. (You may need to
SELECT * remove the '====' lines in 1.* releases.) Be warned that a debug level
FROM employees ['July 24, 1996 09:00:00'] greater than one generates large log files in 1.* releases.
<P>
displays employee rows in the table at the specified time. You can You can actuall run the postgres backend from the command line, and type
specify intervals like [date,date], [date,], [,date], or [,]. This your SQL statement directly. This is recommended ONLY for debugging
last option accesses all rows that ever existed. purposes. Note that a newline terminates the query, not a semicolon. If
you have compiled with debugging symbols, you can perhaps use a debugger
INSERTed rows get a timestamp too, so rows that were not in the table to see what is happening. Because the backend was not started from the
at the desired time will not appear. postmaster, it is not running in an identical environment and
locking/backend interaction problems may not be duplicated. Some
Vacuum removes rows that are no longer current. This time-warp feature operating system can attach to a running backend directly to diagnose
is used by the engine for rollback and crash recovery. Expiration problems.
times can be set with purge. <P>
The postgres program has a -s, -A, -t options that can be very usefull
In 6.0, once a table is vacuumed, the creation time of a row may be for debugging and performance measurements.
incorrect, causing time-traval to fail. <P>
The EXPLAIN command (see this FAQ) allows you to see how PostgreSQL is
The time-travel feature will be removed in 7.0. iterpreting your query.
<P>
3.38) How do I tune the database engine for better performance? <H3><a name="3.37">3.37</a>) What is an oid? What is a tid?</H3>
<P>
There are two things that can be done. You can use Openlink's option Oids are Postgres's answer to unique row ids or serial columns. Every
to disable fsync() by starting the postmaster with a '-o -F' option. row that is created in Postgres gets a unique oid. All oids generated
This will prevent fsync()'s from flushing to disk after every by initdb are less than 16384 (from backend/access/transam.h). All
transaction. post-initdb (user-created) oids are equal or greater that this. All
these oids are unique not only within a table, or database, but unique
You can also use the postmaster -B option to increase the number of within the entire postgres installation.
shared memory buffers shared among the backend processes. If you make <P>
this parameter too high, the process will not start or crash Postgres uses oids in its internal system tables to link rows in
unexpectedly. Each buffer is 8K and the defualt is 64 buffers. separate tables. These oids can be used to identify specific user rows
and used in joins. It is recommended you use column type oid to store
3.39) What debugging features are available in PostgreSQL? oid values. See the sql(l) manual page to see the other internal
columns.
PostgreSQL has several features that report status information that <P>
can be valuable for debugging purposes. Tids are used to indentify specific physical rows with block and offset
values. Tids change after rows are modified or reloaded. They are used
First, by compiling with DEBUG defined, many assert()'s monitor the by index entries to point to physical rows. They can not be accessed
progress of the backend and halt the program when something unexpected through sql.
occurs. <P>
<H3><a name="3.38">3.38</a>) What is the meaning of some of the terms
Both postmaster and postgres have several debug options available. used in Postgres?</H3>
First, whenever you start the postmaster, make sure you send the <P>
standard output and error to a log file, like: Some of the source code and older documentation use terms that have more
common usage. Here are some:
<UL>
cd /usr/local/pgsql <LI> row, record, tuple
./bin/postmaster >server.log 2>&1 & <LI> attribute, field, column
<LI> table, class
This will put a server.log file in the top-level PostgreSQL directory. <LI> retrieve, select
This file can contain useful information about problems or errors <LI> replace, update
encountered by the server. Postmaster has a -d option that allows even <LI> append, insert
more detailed information to be reported. The -d option takes a number <LI> oid, serial value
1-3 that specifies the debug level. The query plans in a verbose debug <LI> portal, cursor
file can be formatted using the 'indent' program. (You may need to <LI> range variable, table name, table alias
remove the '====' lines in 1.* releases.) Be warned that a debug level </UL>
greater than one generates large log files in 1.* releases. <P>
Please let me know if you think of any more.
You can actuall run the postgres backend from the command line, and <P>
type your SQL statement directly. This is recommended ONLY for <H3><a name="3.39">3.39</a>) What is Genetic Query Optimization?</H3>
debugging purposes. Note that a newline terminates the query, not a <P>
semicolon. If you have compiled with debugging symbols, you can The GEQO module in PostgreSQL is intended to solve the query
perhaps use a debugger to see what is happening. Because the backend optimization problem of joining many tables by means of a Genetic
was not started from the postmaster, it is not running in an identical Algorithm (GA). It allows the handling of large join queries through
environment and locking/backend interaction problems may not be non-exhaustive search.
duplicated. Some operating system can attach to a running backend <P>
directly to diagnose problems. For further information see README.GEQO &lt;utesch@aut.tu-freiberg.de&gt;.
<P>
The postgres program has a -s, -A, -t options that can be very usefull <H3><a name="3.40">3.40</a>) I am running Solaris and my dates
for debugging and performance measurements. display wrong. Why?</H3>
<P>
The EXPLAIN command (see this FAQ) allows you to see how PostgreSQL is There was a bug in 6.0 that caused this problem under Solaris with -O2
iterpreting your query. optimization.
Upgrade to 6.1.
3.40) What is an oid? What is a tid? <P>
<H3><a name="3.41">3.41</a>) How do I enable more than 32 concurrent
Oids are Postgres's answer to unique row ids or serial columns. Every backends?</H3>
row that is created in Postgres gets a unique oid. All oids generated <P>
by initdb are less than 16384 (from backend/access/transam.h). All Edit include/storage/sinvaladt.h, and change the value of MaxBackendId.
post-initdb (user-created) oids are equal or greater that this. All In the future, we plan to make this a configurable prameter.
these oids are unique not only within a table, or database, but unique <P>
within the entire postgres installation. <HR>
<H2> Section 4: Extending PostgreSQL
Postgres uses oids in its internal system tables to link rows in </H2>
separate tables. These oids can be used to identify specific user rows <P>
and used in joins. It is recommended you use column type oid to store <H3><a name="4.1">4.1</a>) I wrote a user-defined function and when I run it
oid values. See the sql(l) manual page to see the other internal in psql, it dumps core.</H3>
columns. <P>
The problem could be a number of things. Try testing your user-defined
Tids are used to indentify specific physical rows with block and function in a stand alone test program first. Also, make sure you are
offset values. Tids change after rows are modified or reloaded. They not sending elog NOTICES when the front-end is expecting data, such as
are used by index entries to point to physical rows. They can not be during a type_in() or type_out() functions
accessed through sql. <P>
<H3><a name="4.2">4.2</a>) I get messages of the type
3.41) What is the meaning of some of the terms used in Postgres? NOTICE:PortalHeapMemoryFree: 0x402251d0 not in alloc set!</H3>
<P>
Some of the source code and older documentation use terms that have You are pfree'ing something that was not palloc'ed. When writing
more common usage. Here are some: user-defined functions, do not include the file "libpq-fe.h". Doing so
* row, record, tuple will cause your palloc to be a malloc instead of a free. Then, when the
* attribute, field, column backend pfrees the storage, you get the notice message.
* table, class <P>
* retrieve, select <H3><a name="4.3">4.3</a>) I've written some nifty new types and functions for
* replace, update PostgreSQL.</H3>
* append, insert <P>
* oid, serial value Please share them with other PostgreSQL users. Send your extensions to
* portal, cursor mailing list, and they will eventually end up in the contrib/
* range variable, table name, table alias subdirectory.
<P>
Please let me know if you think of any more. <H3><a name="4.4">4.4</a>) How do I write a C function to return a tuple?</H3>
<P>
3.42) What is Genetic Query Optimization? This requires extreme wizardry, so extreme that the authors have not
ever tried it, though in principle it can be done. The short answer is
The GEQO module in PostgreSQL is intended to solve the query ... you can't. This capability is forthcoming in the future.
optimization problem of joining many tables by means of a Genetic <P>
Algorithm (GA). It allows the handling of large join queries through <HR>
non-exhaustive search. <H2> Section 5: Bugs
</H2>
For further information see README.GEQO <utesch@aut.tu-freiberg.de>. <P>
<H3><a name="5.1">5.1</a>) How do I make a bug report?</H3>
3.43) I am running Solaris and my dates display wrong. Why? <P>
Check the current FAQ at <a
Gcc under Solaris has an bug when compiled with optimization level 2. href="http://postgreSQL.org">http://postgreSQL.org</a>
Edit 'configure', and change -O2 to -O, and recompile. If 'configure' <P>
does not exist in your top level source directory, make the change to Also check out our ftp site <a
Makefile.global. href="ftp://ftp.postgreSQL.org/pub">ftp://ftp.postgreSQL.org/pub</a> to
see if there is a more recent PostgreSQL version.
_________________________________________________________________ <P>
You can also fill out the "bug-template" file and send it to:
Section 4: Extending PostgreSQL <UL>
<LI> <a href="mailto:bugs@postgreSQL.org">bugs@postgreSQL.org</a>
4.1) I wrote a user-defined function and when I run it in psql, it dumps </UL>
core. <P>
This is the address of the developers mailing list.
The problem could be a number of things. Try testing your user-defined </BODY>
function in a stand alone test program first. Also, make sure you are </HTML>
not sending elog NOTICES when the front-end is expecting data, such as
during a type_in() or type_out() functions
4.2) I get messages of the type NOTICE:PortalHeapMemoryFree: 0x402251d0 not
in alloc set!
You are pfree'ing something that was not palloc'ed. When writing
user-defined functions, do not include the file "libpq-fe.h". Doing so
will cause your palloc to be a malloc instead of a free. Then, when
the backend pfrees the storage, you get the notice message.
4.3) I've written some nifty new types and functions for PostgreSQL.
Please share them with other PostgreSQL users. Send your extensions to
mailing list, and they will eventually end up in the contrib/
subdirectory.
4.4) How do I write a C function to return a tuple?
This requires extreme wizardry, so extreme that the authors have not
ever tried it, though in principle it can be done. The short answer is
... you can't. This capability is forthcoming in the future.
_________________________________________________________________
Section 5: Bugs
5.1) How do I make a bug report?
Check the current FAQ at http://postgreSQL.org
Also check out our ftp site ftp://ftp.postgreSQL.org/pub to see if
there is a more recent PostgreSQL version.
You can also fill out the "bug-template" file and send it to:
* bugs@postgreSQL.org
This is the address of the developers mailing list.
doc/TODO
View file @ 9b22846e
...@@ -122,6 +122,8 @@ add pg_type attribute to identify types that need length (bpchar, varchar) ...@@ -122,6 +122,8 @@ add pg_type attribute to identify types that need length (bpchar, varchar)
add UNIQUE capability to non-btree indexes add UNIQUE capability to non-btree indexes
make pg_dumpall preserve table ownership, not just database ownership make pg_dumpall preserve table ownership, not just database ownership
make large objects have their own reltype make large objects have their own reltype
make number of backends a config parameter, storage/sinvaladt.h:MaxBackendId
certain indexes will not shrink, i.e. oid indexes with many inserts
PERFORMANCE PERFORMANCE
----------- -----------
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment